Particle-based methods are powerful tools for simulating systems with complex geometries, free surfaces, and tightly coupled multiphysics. We present TrixiParticles.jl, a high-performance, open-source framework in Julia designed to make these methods accessible and extensible. TrixiParticles.jl supports a variety of modern simulation techniques, featuring multiple Smoothed Particle Hydrodynamics (SPH) schemes (WCSPH, EDAC, IISPH) and solid mechanics models (TLSPH, DEM) for robust fluid-structure interaction. Its core design philosophy leverages Julia's composable nature, enabling users to express complex setups and couple particle systems directly in high-level code. This facilitates rapid prototyping and extension without sacrificing performance. The framework is built for speed, incorporating an optimized neighbor search, native multithreading, and vendor-agnostic GPU execution (NVIDIA, AMD, Apple) through KernelAbstractions.jl. Flexible time integration is handled by the OrdinaryDiffEq.jl library. A comprehensive workflow is supported by utilities for particle sampling from STL/ASC files and VTK output for visualization. To promote education and reproducible science, the framework includes a rich set of validation cases and examples. This poster introduces the design of TrixiParticles.jl, summarizes its key features, and highlights compelling validation and benchmark results.